Meaning distinctions in terminology research

Information and argument structures in Sociology research abstracts

The subject of the research is the search for the optimal method of analyzing deep structures using frame semantics. The study of semantic roles, similarities and differences in the approaches of both C. Fillmore and B. Levin – M. R. Hovav make it possible to analyze the structure of a sentence in more detail and accurately, identify deep cases and determine semantic relations between words. The study of these aspects is key to understanding language constructs and their interpretation. The study of various approaches makes it possible to identify both common features and unique features, which is key for a complete understanding of language constructions. An interest in text analysis in the field of artificial intelligence, machine learning and computational linguistics, and an understanding of the semantic relationships between words will help create more accurate and efficient text processing algorithms. One of the research methods is the semantic analysis of sentences based on corpus data. This method includes the study of various linguistic constructions in the context of their use in real texts, which allows us to identify common patterns and rules for the use of these linguistic units in different situations. The scientific novelty of the study lies in the fact that the authors have determined the similarity of the approaches on how to understand the surface and deep structures of language of Ch. Fillmore and B. Levin and M. Rappaport. Their work, despite differences in methodology and terminology, together allow for in-depth investigation of the relationship between the meanings of verbs and the structure of arguments. As a result of the study, the natural relationships between deep cases and semantic roles in sentences of various types are revealed, and key points that need to be taken into account when analyzing deep structures for a more accurate definition of the semantic roles of arguments are highlighted: frame semantics and thematic grids. Disagreements and alternative points of view contribute to the constant development and improvement of linguistic theories. Such debates eventually lead to a deeper understanding of the implementation of the arguments and open up opportunities for further research in this area. Both C. Fillmore and B. Levin and M. Rappaport have made significant contributions to understanding the surface and deep structures of language, although their approaches and terminology may differ.

We present a brief history and overview of statistical methods in frame-semantic parsing – the automatic analysis of text using the theory of frame semantics. We discuss how the FrameNet lexicon and frameannotated datasets have been used by statistical NLP researchers to build usable, state-of-the-art systems. We also focus on future directions in frame-semantic parsing research, and discuss NLP applications that could benefit from this line of work. 1 Frame-Semantic Parsing Frame-semantic parsing has been considered as the task of automatically finding semantically salient targets in text, disambiguating their semantic frame representing an event and scenario in discourse, and annotating arguments consisting of words or phrases in text with various frame elements (or roles). The FrameNet lexicon (Baker et al., 1998), an ontology inspired by the theory of frame semantics (Fillmore, 1982), serves as a repository of semantic frames and their roles. Figure 1 depicts a sentence with three evoked frames for the targets “million”, “created” and “pushed” with FrameNet frames and roles. Automatic analysis of text using framesemantic structures can be traced back to the pioneering work of Gildea and Jurafsky (2002). Although their experimental setup relied on a primitive version of FrameNet and only made use of “exemplars” or example usages of semantic frames (containing one target per sentence) as opposed to a “corpus” of sentences, it resulted in a flurry of work in the area of automatic semantic role labeling (Marquez et al., 2008). However, the focus of semantic role labeling (SRL) research has mostly been on PropBank (Palmer et al., 2005) conventions, where verbal targets could evoke a “sense” frame, which is not shared across targets, making the frame disambiguation setup different from the representation in FrameNet. Furthermore, it is fair to say that early research on PropBank focused primarily on argument structure prediction, and the interaction between frame and argument structure analysis has mostly been unaddressed (Marquez et al., 2008). There are exceptions, where the verb frame has been taken into account during SRL (Meza-Ruiz and Riedel, 2009; Watanabe et al., 2010). Moreoever, the CoNLL 2008 and 2009 shared tasks also include the verb and noun frame identification task in their evaluations, although the overall goal was to predict semantic dependencies based on PropBank, and not full argument spans (Surdeanu et al., 2008; Hajic et al., 2009). The SemEval 2007 shared task (Baker et al., 2007) attempted to revisit the frame-semantic analysis task based on FrameNet. It introduced a larger FrameNet lexicon (version 1.3), and also a larger corpus with full-text annotations compared to prior work, with multiple targets annotated per sentence. The corpus allowed words and phrases with noun, verb, adjective, adverb, number, determiner, conjunction and preposition syntactic categories to serve as targets and evoke frames, unlike any other single dataset; it also allowed targets from different syntactic categories share frames, and therefore roles. The repository of semantic role types was also much richer than PropBankstyle lexicons, numbering in several hundreds. Most systems participating in the task resorted to a cascade of classifiers and rule-based modules: identifying targets (a non-trivial subtask), disambiguating frames, identifying potential arguments, and then labeling them with roles. The system described by Johansson and Nugues (2007) performed the best in this shared task. Next, we focus on its performance, and subsequent improvements made by the research community on this task.

We present a system for identifying the semantic relationships, or semantic roles, filled by constituents of a sentence within a semantic frame. Given an input sentence and a target word and frame, the system labels constituents with either abstract semantic roles, such as Agent or Patient, or more domain-specific semantic roles, such as Speaker, Message, and Topic. The system is based on statistical classifiers trained on roughly 50,000 sentences that were hand-annotated with semantic roles by the FrameNet semantic labeling project. We then parsed each training sentence into a syntactic tree and extracted various lexical and syntactic features, including the phrase type of each constituent, its grammatical function, and its position in the sentence. These features were combined with knowledge of the predicate verb, noun, or adjective, as well as information such as the prior probabilities of various combinations of semantic roles. We used various lexical clustering algorithms to generalize across possible fillers of roles. Test sentences were parsed, were annotated with these features, and were then passed through the classifiers. Our system achieves 82% accuracy in identifying the semantic role of presegmented constituents. At the more difficult task of simultaneously segmenting constituents and identifying their semantic role, the system achieved 65% precision and 61% recall. Our study also allowed us to compare the usefulness of different features and feature combination methods in the semantic role labeling task. We also explore the integration of role labeling with statistical syntactic parsing and attempt to generalize to predicates unseen in the training data.

The meaning of a predicate, especially a verb, may be characterized via the relations that its arguments bear to it. Semantic roles—also known as thematic relations, theta roles, participant roles, and deep cases—are labels for certain recurring predicate-argument relations. They have proved attractive because they provide a way of representing commonalities across different uses of the same predicate or across uses of distinct but semantically related predicates that may be obscured because arguments with certain semantic roles may have various syntactic realizations. Thus they provide a level of abstraction for the statement of generalizations concerning a variety of linguistic phenomena. In particular, argument realization generalizations are often stated over a thematic hierarchy, a ranking of semantic roles. However, semantic roles have not lived up to their initial promise. It has proved impossible to find a small set of roles that can be applied across all verbs in a language, let alone across languages. Yet this desideratum must be met if semantic roles are to figure effectively in accounts of linguistic phenomena. Further, some generalizations involving semantic roles seem to require reference to coarse-grained roles, whereas others require reference to fine-grained roles. Moreover, reliable diagnostics are difficult to identify even for the roles cited most often. Although these difficulties have led some researchers to reject semantic roles, others have taken alternative approaches, including the use of generalized semantic roles, which are inspired by the notion of prototype, with no single property being necessary or sufficient for an argument to bear such a role. Despite these drawbacks, semantic roles continue to be useful in stating linguistic generalizations, and so descriptive, if not theoretical, uses of semantic role labels persist across subfields, including language acquisition, psycholinguistics, and neurolinguistics. Furthermore, semantic roles are useful in natural language processing. Since semantic roles have been implicated in phenomena involving argument structure, the separate Oxford Bibliographies article Argument Structure should be consulted for additional relevant resources. Acknowledgments: For discussion of the material in this article, the author is grateful to Scott Grimm, Chris Manning, Malka Rappaport Hovav, and two reviewers, as well as the students in her autumn 2012 lexical semantics class.

Beginning with an overview of the theory of Frame Semantics as developed by Charles Fillmore and colleagues, this article details the annotation of English sentences by the FrameNet Project based on this theory. Fillmore’s lexical semantics theory asserts that the meanings of most words are understood via the semantic frames they evoke; e.g. arrest, apprehend, apprehension, bust, and nab can all evoke the Arrest frame, with its associated frame-specific semantic roles: Suspect, Authorities, Offense, and Charges. Thus, They were busted for shoplifting by three plainclothes policemen would be labeled to show that bust is the frame-evoking expression, they fills the Suspect role, for shoplifting is the Offense, and by three plainclothes policemen represents the Authorities. Combining multiple annotations of this type creates a picture of the valence (valency) patterns of the lexical unit (word sense) and the semantic frame. The resulting database contains more than 200,000 manual annotations of 13,500 lexical units in 1,200 semantic frames. Expanding from the original goal of lexicography, the team has annotated a number of texts “fully”, i.e. labeling all the frame-evoking elements and the phrases that fill their semantic roles, providing a rich representation of the lexical semantics of the entire text. Automatic semantic role labeling systems trained on FrameNet can label a wide range of texts with increasing accuracy for NLP research and applications. The author describes current limitations and possible extensions of this methodology and how the practice of manual annotation informs the development of the theory.

We present results on a novel hybrid semantic SMT model that incorporates the strengths of both semantic role labeling and phrase-based statistical machine translation. The approach avoids major complexity limitations via a two-pass architecture. The first pass is performed using a conventional phrase-based SMT model. The second pass is performed by a re-ordering strategy guided by shallow semantic parsers that produce both semantic frame and role labels. Evaluation on a Wall Street Journal newswire genre test set showed the hybrid model to yield an improvement of roughly half a point in BLEU score over a strong pure phrase-based SMT baseline -- to our knowledge, the first successful application of semantic role labeling to SMT.

Reviewed by: Lexical semantics, syntax, and event structure Lisa Levinson Lexical semantics, syntax, and event structure. Ed. by Malka Rappaport Hovav, Edit Doron, and Ivy Sichel. (Oxford studies in theoretical linguistics.) New York: Oxford University Press, 2010. Pp. xvi, 401. ISBN 9780199544332. $54.95. This volume consists of a diverse set of papers presented at a workshop held in 2006 in honor of Anita Mittwoch for the occasion of her eightieth birthday. The workshop was entitled 'Syntax, [End Page 420] lexicon, and event structure', and was themed to relate to Mittwoch's work on event temporality and its relation to syntax and the lexicon. One can view the sequence of three parts as building a path from analyzing the atomic elements of event descriptions up to investigating the interaction of larger constituents at the phrasal level. The variety of the articles accurately reflects the range of empirical issues and theoretical approaches in the field. The introduction to the volume makes an impressive attempt at offering a background for the wide range of chapters that follow. It includes a section that lays out the general path of the volume and another that summarizes each of the articles, occasionally relating them to one another. Given the diversity of the empirical coverage of the chapters, there is not much comparison of analyses of the same phenomena, just as in the chapters themselves there is not much referencing of other chapters. The editors conclude the chapter with a tribute to Anita Mittwoch, reviewing her academic contributions and drawing connections between her work and that of several of the contributors to the volume. Part 1, 'Lexical representation', is relevant in that such representations play a role in shaping event descriptions. In Ch. 2, 'Reflections on manner/result complementarity', Malka Rappaport Hovav and Beth Levin argue that event structures are constrained such that they may have roots that are manner or result, but not both at the same time. By 'root' they mean the core lexical content of the word, as divorced from structural properties associated with it in a specific event structure or syntactic context. They argue that their explanation for this complementarity provides insight into the nature of event structures more broadly. In Ch. 3, 'Verbs, constructions, and semantic frames', Adele Goldberg argues that no such manner/result complementarity exists as a constraint on verbs or event structure. She argues from a construction-grammar perspective that the primary constraint on possible event structures encoded by verbs is what she calls the conventional frame constraint. This constraint does not depend on root types in the way that Rappaport Hovav and Levin's analysis does; rather Goldberg posits that a verb meaning is possible so long as it describes events that form a coherent semantic frame. Like Rappaport Hovav and Levin, Nomi Erteschik-Shir and Tova Rapoport, in Ch. 4, 'Contact and other results', make use of root types to account for the distribution of roots across different event structures and argument structures. Their approach is more syntactically oriented than that of Rappaport Hovav and Levin, in the spirit of Ken Hale and Samuel Jay Keyser. More specifically, they propose atom theory, whereby alternations with contact verbs arise from a combination of the meaning of the root and the interpretation of the structure in which the root is embedded. The final paper in Part 1, Ch. 5, 'The lexical encoding of idioms' by Martin Everaert, does not focus directly on the types (or lack thereof) of roots. Instead, Everaert approaches the lexicon from the perspective of idiom interpretation. Part 2, 'Argument structure and the compositional construction of predicates', includes papers that focus on syntactically and semantically larger or 'higher' elements, such as heads introducing external arguments and causativity. In Ch. 6, 'The emergence of argument structure in two new sign languages', Irit Meir demonstrates, based on Israeli Sign Language, that marking argument structure for polyadic predicates with verb agreement does not require an earlier stage of the language where it is indicated with word order. This is a new finding, as work on the development of spoken languages has suggested...

The article deals with the problem of semantic roles’ distribution in the frame VIRUS designated by the term virus in the computer virology discourse. The study is conducted in the framework of the cognitive discursive paradigm of modern linguistics and comprises the following linguistic approaches to the studies of language for special purposes: cognitive terminology, frame-based terminology, and frame semantics. The article gives a brief overview of the development of computer virology with reference to mental framing of the key aspects in the field. A frame is considered as part of context and a situation model representing a real-life event. Ch. Fillmore’s frame semantics and the identification of deep cases or semantic roles are used as the main method of data analyses. We have analyzed the most typical plans of semantic roles’ distribution in the frame VIRUS. The semantic roles of the frame VIRUS include Agent, Counteragent, Object, Addressee, Patient, Result, and Instrument. It has been found that besides the most distinctive distribution of semantic roles in the frame VIRUS, showing that the malicious program is conceptualized as the aggressor and a computer or its user as a victim, which correspond to the roles of the Agent and the Patient, there might be the frames with a virus in the role of the Counteragent, the Object, the Result, the Instrument, the Patient, and even the Place. We come to the conclusion that analysis of the roles’ distribution helps to determine relations among the event participants and the way the situation is conceptualized and represented in the form of mental models in human minds. In the Discussion section we illustrate the occurrence of typical semantic frames in popular culture, namely in films and television programs. With these we prove the relevance of the approaches and methods chosen to reveal the peculiarities of conceptualization in special areas and connection synergies between language, thought, and communication.

Reviewed by: Proto-properties and grammatical encoding: A correspondence theory of argument selection by Farrell Ackerman and John Moore Kleanthes K. Grohmann Proto-properties and grammatical encoding: A correspondence theory of argument selection. By Farrell Ackerman and John Moore. (Stanford monographs in linguistics.) Stanford, CA: CSLI Publications, 2001. Pp. ix, 197. ISBN 1575861666. $18.95. Ackerman and Moore develop an interesting theory with wide-ranging implications for morphosemantic alternations within David Dowty’s proto-role approach to thematic roles (‘Thematic roles and argument selection’, Language 67.547–619, 1991), which treats thematic roles as prototype categories. The authors follow lexical unification-based assumptions embedded within Ray Jackendoff’s representational modularity-architecture of the language faculty (The architecture of the language faculty, Cambridge, ma: mit Press, 1997). Ch. 1 offers an ‘Introduction’ (1–13) to basic linking problems in argument structure, provides an overview of the book, and presents the model of correspondence theory which is employed throughout. This particular linking theory aims at accounting for both grammatical function and case-marking alternations by appealing to the same explanatory apparatus. Couched in a useful overview of semantic roles in argument structure, the next two chapters present Dowty’s proposal in detail. a short Ch. 2 deals with ‘Grammatical theory and semantic roles’ (15–25), [End Page 766] while the more elaborate Ch. 3 presents ‘a proto-role proposal: Syntagmatic argument selection’ (27–59), where the syntagmatic selection principle (ssp) is A&M’s revision of Dowty’s argument selection principle. Within correspondence theory, coarguments of a given predicate are associated with proto-properties; the most proto-agentive argument becomes the subject and the most proto-patientive, object. Ch. 4 introduces ‘Paradigmatic argument selection’ (61–82), which involves an extension of the ssp to paradigms. The paradigmatic selection principle (psp) formalizes A&M’s claim that a single argument competes for alternative encodings across related predicates (a&m illustrate this with psych-predicates and causative formation in Spanish). The next two chapters make up the empirical core of the monograph, exploring several of the predictions and conceptual issues advanced in Ch. 4. Ch. 5 introduces ‘Bounding entities’ (83–139); an addition (distinct from the incremental theme) to the list of proto-patient properties, bounding entities are proposed alongside a detailed discussion of aspectual issues in Estonian (as well as the related Finnish and the less related Scots Gaelic). In Ch. 6, ‘Subject alternations’ (141–68), a&m concentrate on subject-oblique alternations in Polish and Russian along with a discussion of oblique (dative) subject phenomena crosslinguistically, especially in Hindi and Marathi. In Ch. 7, ‘Conclusion’ (169–77), a&m summarize the main results of their study, concluding that their proto-property approach containing syntagmatic and paradigmatic selection accounts for both the grammatical function status and the case-marking encoding of arguments. They also suggest an extension of paradigmatic selection to new data domains and end with a brief comparison of their approach to other linking theories. The bibliography (179–90) is followed by an index (191–97). This excellent monograph is recommended to anyone interested in the prototype approach to thematic relations and argument structure, but, while well written and illustrated throughout, it is clearly not intended for the uninitiated. [End Page 767] Kleanthes K. Grohmann University of Cyprus Copyright © 2005 Linguistic Society of America

Resources such as FrameNet, which provide sets of semantic frame definitions and annotated textual data that maps into the evoked frames, are important for several NLP tasks. However, they are expensive to build and, consequently, are unavailable for many languages and domains. Thus, approaches able to induce semantic frames in an unsupervised manner are highly valuable. In this paper we approach that task from a network perspective as a community detection problem that targets the identification of groups of verb instances that evoke the same semantic frame and verb arguments that play the same semantic role. To do so, we apply a graph-clustering algorithm to a graph with contextualized representations of verb instances or arguments as nodes connected by edges if the distance between them is below a threshold that defines the granularity of the induced frames. By applying this approach to the benchmark dataset defined in the context of SemEval 2019, we outperformed all of the previous approaches to the task, achieving the current state-of-the-art performance.

The impact of frame semantic enrichment of texts on the task of factoid question answering (QA) is studied in this paper. In particular, we consider different techniques for answer processing with frame semantics: the level of semantic class identification and role assignment to texts, and the fusion of frame semantic‐based answer‐processing approaches with other methods used in the Text REtrieval Conference (TREC). The impact of each of these aspects on the overall performance of a QA system is analyzed in this paper. The TREC 2004 and TREC 2006 factoid question sets were used for the experiments. These demonstrate that the exploitation of encapsulated frame semantics in FrameNet in a shallow semantic parsing process can enhance answer‐processing performance in factoid QA systems. This improvement is dependent on the level of semantic annotation, the frame semantic alignment method, and the method of fusing frame semantic‐based answer‐processing models with other existing models. A more comprehensively annotated environment with all different part‐of‐speech target predicates provides a higher chance of correct factoid answer retrieval where semantic alignment is based on both semantic classes and a relaxed set of semantic roles for answer span identification. Our experiments on fusion techniques of frame semantic‐based and entity‐based answer‐processing models show that merging answer lists with respect to their scores and redundancy by exploiting a fusion function leads to a more effective overall factoid QA system compared to the use of individual models.

This study examines the semantic role analysis of Korean symmetric constructions, focusing on how arguments in such constructions share and organize their semantic roles. Symmetric constructions are characterized by the interchangeability of arguments without altering the truth value of the sentence. While previous studies have often analyzed the comitative arguments in symmetric constructions as companions(동반) or counterparts(상대), these approaches fail to account for the full range of semantic roles the arguments may fulfill. This paper proposes that each argument can simultaneously bear multiple roles, so-called bundled semantic roles, at once, based on their bidirectional participation in the event. The paper also explores the distinction between strongly and weakly symmetric predicates depending on their degree of symmetry and considers how the integration of syntactic constituents affects role assignment. Furthermore, it classifies symmetric constructions into six semantic subtypes—combination, separation, relation, comparison, communication, and replacement/exchange—and analyzes how specific semantic roles are realized in each case. By refining the role structure of symmetric constructions and introducing nuanced criteria for analysis, this study contributes a practical framework for future research on argument structure and semantic roles in Korean.

Determining the semantic role of sentence constituents is a key task in determining sentence meanings lying behind a veneer of variant syntactic expression. We present a model of natural language generation from semantics using the FrameNet semantic role and frame ontology. We train the model using the FrameNet corpus and apply it to the task of automatic semantic role and frame identification, producing results competitive with previous work (about 70% role labeling accuracy). Unlike previous models used for this task, our model does not assume that the frame of a sentence is known, and is able to identify null-instantiated roles, which commonly occur in our corpus and whose identification is crucial to natural language interpretation.KeywordsMachine TranslationSemantic RoleStatistical Machine TranslationLexical UnitNatural Language GenerationThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Frame semantic parsing is a fundamental NLP task, which consists of three subtasks: frame identification, argument identification and role classification. Most previous studies tend to neglect relations between different subtasks and arguments and pay little attention to ontological frame knowledge defined in FrameNet. In this paper, we propose a Knowledge-guided Incremental semantic parser with Double-graph (KID). We first introduce Frame Knowledge Graph (FKG), a heterogeneous graph containing both frames and FEs (Frame Elements) built on the frame knowledge so that we can derive knowledge-enhanced representations for frames and FEs. Besides, we propose Frame Semantic Graph (FSG) to represent frame semantic structures extracted from the text with graph structures. In this way, we can transform frame semantic parsing into an incremental graph construction problem to strengthen interactions between subtasks and relations between arguments. Our experiments show that KID outperforms the previous state-of-the-art method by up to 1.7 F1-score on two FrameNet datasets. Our code is availavle at https://github.com/PKUnlp-icler/KID.